Wind Turbine Safety Stop Prediction for Controlled Derating
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wind turbines face structural damage and excessive wear due to sudden safety stops, which are triggered to prevent severe incidents but can cause damage themselves and require manual restarts, disrupting operation and increasing maintenance needs.
Innovation Solution
An electronic device using machine learning models predicts upcoming safety stops by analyzing wind turbine and sensor data, allowing for controlled stops or derating to prevent safety stops, enabling automatic or remote restarts and reducing wear.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a safety stop is triggered to prevent severe incidents, then the wind turbine is protected from structural damage, but the safety stop itself causes structural damage and excessive wear
Solution Approach 1:
The system performs preliminary analysis of wind turbine data and sensor data to predict an upcoming safety stop before it occurs. This allows the main control system to initiate a controlled stop in advance, preventing the harmful rapid deceleration of a safety stop while still protecting the turbine from severe incidents through proactive intervention
Solution Approach 2:
The prediction system acts as an intermediary between the safety control system and the main control system. It analyzes data to foresee potential safety stops and communicates this information to the main control system, which then implements a controlled stop as an intermediate solution that avoids the harmful effects of direct safety stops
2Reliability
If a safety stop is triggered, then severe incidents are prevented, but manual restart is required which disrupts operation
Solution Approach 1:
By predicting an upcoming safety stop before it occurs, the system allows the main control system to execute a controlled stop and prepare for automatic restart. This preliminary action ensures that when the stop occurs, the turbine can be restarted automatically or remotely without requiring manual intervention, thus maintaining operational continuity and productivity
Solution Approach 2:
The system continuously monitors wind turbine data and sensor data, providing feedback to the prediction model. This feedback loop enables the system to learn from patterns and improve prediction accuracy, allowing for more effective proactive control actions that maintain uptime while preventing severe incidents
3Productivity
If a controlled stop is implemented instead of safety stop, then automatic restart is possible improving uptime, but the prediction system requires complex data analysis
Solution Approach 1:
The prediction system leverages existing wind turbine data and sensor data that are already collected for normal operation and safety monitoring. By utilizing these existing data sources for prediction purposes, the system achieves multi-functionality without requiring additional sensors or data collection infrastructure, thus limiting the increase in device complexity
Solution Approach 2:
The machine learning model is trained using historical wind turbine data and sensor data from the same system it serves. The system essentially analyzes its own operational data to generate predictions, eliminating the need for external complex analysis systems and reducing overall device complexity while maintaining predictive capability
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed is a method, performed by an electronic device, for control of operation of a wind turbine. The method comprises obtaining wind turbine data indicative of one or more alerting events of the wind turbine, wherein the wind turbine data comprises warning data indicative of a warning of the wind turbine, and/or alarm data indicative of an alarm state of the wind turbine. The method comprises obtaining sensor data from a plurality of sensors of the wind turbine indicative of operating conditions associated with the wind turbine. The method comprises predicting, based on the wind turbine data and the sensor data, an upcoming safety stop by applying a machine learning model to the wind turbine data and the sensor data. The method may comprise providing, based on the predicted upcoming safety stop, control data indicative of a controlled stop or a derating to be performed by the wind turbine.